Exec (new process or new thread?) to continue

ruby-talk
1

Hi,

I have fired off an exe with the kernal method exec

loader_app = "#{Dir.getwd}/PublishUi.exe"
exec(loader_app)

The exec method seems to hold up the process

so

exec(loader_app)
puts "we got here"

Will not print "we got here" till I close the app.

I am not sure whether I should use a new thread or a new process here,
or whether there is an alternative.

Thanks

Aidy

2

IO.popen(an_exe)

···

On Feb 24, 2009, at 8:19 AM, aidy <aidy.lewis@googlemail.com> wrote:

Hi,

I have fired off an exe with the kernal method exec

loader_app = "#{Dir.getwd}/PublishUi.exe"
exec(loader_app)

The exec method seems to hold up the process

so

exec(loader_app)
puts "we got here"

Will not print "we got here" till I close the app.

I am not sure whether I should use a new thread or a new process here,
or whether there is an alternative.

Thanks

Aidy

3

aidy wrote:

Hi,

I have fired off an exe with the kernal method exec

loader_app = "#{Dir.getwd}/PublishUi.exe"
exec(loader_app)

The exec method seems to hold up the process

so

exec(loader_app)
puts "we got here"

Will not print "we got here" till I close the app.

I am not sure whether I should use a new thread or a new process here,
or whether there is an alternative.

Thanks

Aid

exec doesn't hold up the process, it replaces it completely[1]. Your second example will never print "we got here".

Assuming you do not need to communicate with, wait on, or know anything about the process you are starting, you can use fork[2] and exec this way:

fork { exec load_app }

-Justin

[1] module Kernel - RDoc Documentation
[2] module Kernel - RDoc Documentation

4

aidy wrote:

Hi,

I have fired off an exe with the kernal method exec

loader_app = "#{Dir.getwd}/PublishUi.exe"
exec(loader_app)

Since you appear to be on windows, you can't fork, but you can do this:

Thread.new do
   loader_app = "#{Dir.getwd}/PublishUi.exe"
   system(loader_app)
end

# continue with other stuff here

This is a fairly cross-platform way to handle it. Of course on windows, you can also do

system "start ..."

and then you don't even need a ruby thread.

···

--
       vjoel : Joel VanderWerf : path berkeley edu : 510 665 3407

5

And with ruby 1.9, Kernel.spawn:
http://ruby-doc.org/core-1.9/classes/Kernel.html#M006071

Cheers,
lasitha.

···

On Tue, Feb 24, 2009 at 6:59 PM, List.rb <list.rb@gmail.com> wrote:

On Feb 24, 2009, at 8:19 AM, aidy <aidy.lewis@googlemail.com> wrote:

The exec method seems to hold up the process
so

exec(loader_app)
puts "we got here"

Will not print "we got here" till I close the app.

IO.popen(an_exe)

6

I like this pattern:

def fire_and_forget(&block)
  pid = fork do
    begin
      yield
    ensure
      Process.exit!
    end
  end
  Process.detach pid
end

A little bit safer then just using 'fork'...

Taken from a post here:
http://www.caboo.se/articles/2006/10/14/premcache-caching-an
d-precaching-with-memcached

Justin Collins wrote:

···

Assuming you do not need to communicate with, wait on, or know anything
about the process you are starting, you can use fork[2] and exec this
way:

fork { exec load_app }

-Justin

[1] module Kernel - RDoc Documentation
[2] module Kernel - RDoc Documentation

--
Posted via http://www.ruby-forum.com/\.